ROTT test on Novus Sealing Uniflon 50 gasket - · PDF file"Uniflon 50" gasket N°...

Click here to load reader

  • date post

    09-Mar-2018
  • Category

    Documents

  • view

    214
  • download

    2

Embed Size (px)

Transcript of ROTT test on Novus Sealing Uniflon 50 gasket - · PDF file"Uniflon 50" gasket N°...

  • Page 1/12

    Toute reproduction partielle susceptible de dnaturer le contenu du prsent document, qu'il s'agisse d'une omission, d'une modification ou d'une adaptation engage la responsabilit du client vis vis du CETIM ainsi que des tiers concerns.

    74, route de la Jonelire - BP 82617 - 44326 Nantes CEDEX 03 - Tl. : 0240373635 - Fax : 0240373699

    N/Rf : 2005/682/AT/CBOL/MTOU

    ROTT test on Novus Sealing "Uniflon 50" gasket

    N : 790730/6J1/a

    Date : December 15th 2005 Test Report

    Destinataire (s) : Mr SMITH Gavin NOVUS SEALING LIMITED HUNSWORTH LANE CLECKHEATON BD 19 3UJ WEST YORKSHIRE (UK)

    Rf. de la demande :

    Purchase order n 0000000535

    lments remis par le demandeur :

    - 6 samples of Novus Sealing Limited Uniflon 50 gasket

  • Le December 15th 2005 AT n 790730/6J1/a Page 2/12

    1. AIM OF TEST

    The aim of ROTT (ROom Temperature Tightness) test is to determine the value of PVRC gasket factors (Gb, a, Gs) at room temperature. ROTT test enables to calculate constants "Gb", "a", "Gs" from leak measurements at various tightening and pressure levels, that ASME will soon use in place of "m" and "y". ROTT test made at CETIM's corresponds to a North-America technology developed by TTRL (Tightness Testing and Research Laboratory) on behalf of the PVRC (Pressure Vessel Research Council).

    As an extension to ROTT test a CRUSH test is performed to determine the maximum allowable stress on the gasket before tightness is damaged.

    2. ROTT + CRUSH TEST PROCEDURE

    2.1 ROTT ASTM DRAFT 9 test procedure

    2.1.1 Test sequence ROTT test procedure includes a series of three (3) loadings and unloadings of the gasket applied in turn stepwise at increasingly high stresses during which the leak is measured at each stress level. A long enough holding time is observed at each measurement for the leak to stabilise. It varies from one (1) minute to five (5) hours. The gas used is helium pressurised at 27.5 and 55 bar. The Figure 2: ROTT Draft 9 test sequence vs time presents the various steps of ROTT test.

    The ROTT test (according to Draft No. 9) can be made following three different procedures depen-ding on the type of gasket to be tested, namely SOFT, STANDARD and HARD. These procedures correspond to the stress levels reached during testing. The SOFT procedure is used with PTFE-based gaskets and also with gaskets specially designed to withstand low stresses (spiral gaskets). In the case of fully metallic gaskets, the test must be made according to the HARD procedure. Table 1: Gasket stress level during a ROTT ASTM Draft9 test shows the stress levels for all these procedures. For this test (ROTT test on Novus Sealing "Uniflon 50" gasket), SOFT procedure has been used. For repeatability purposes, the test is performed at least 2 times. A third test is performed in case of incident during one of the two first tests or in case of incoherence between results of two first tests.

  • Le December 15th 2005 AT n 790730/6J1/a Page 3/12

    Stress applied on the gasket

    Soft Standard Hard Stress level

    Mpa Psi Mpa Psi Mpa Psi

    S1 7.1 1025 7.1 1125 10.6 1540

    S2 20.9 3040 31.4 4560 47.1 6840

    S2.5 29.1 4220 41.6 6325 65.4 9490

    S3 37.2 5390 55.8 8090 83.7 12140

    S3.5 45.3 6575 68.0 9860 102.0 14790

    S4 53.5 7750 80.2 11630 120.3 17450

    S4.5 61.6 8930 92.4 13395 138.5 20095

    S5 69.7 10110 104.2 15160 156.8 22740

    Table 1: Gasket stress level during a ROTT ASTM Draft9 test

    Figure 1: ROTT ASTM Draft9 Gasket stress sequence

  • Le December 15th 2005 AT n 790730/6J1/a Page 4/12

    Figure 2: ROTT Draft 9 test sequence vs time

    Figure 3: ROTT ASTM Draft 9 test sequence Sg vs Dg

  • Le December 15th 2005 AT n 790730/6J1/a Page 5/12

    2.1.2 Leak rate measurement The test consists in measuring the leak on the gasket at various stresses and pressure levels. Three types of stresses are obtained by applying two different pressures in turn (28bar and 55bar). One of the measuring methods below is used depending on the leak level: - flow meter (for a leak > 0.8mg/s) - pressure drop (for a leak ranging from 0.8 to 0.018 mg/s) - helium mass spectrometer (for a leak < 0.018 mg/s)

    2.1.3 Result analysis A dimensionless tightness parameter Tp (Tightness parameter) is calculated on basis of the measurement results. The effective stress on the gasket vs. tightness parameter Tp curve is plotted and used to determine tightness constants "Gb", "a", "Gs". The test must be repeated on the same type of gasket to calculate the constants. The tightness of a gasket proves its capability to control the leak of a pressured joint. Many tightness tests have proven that the measured leak rate and the fluid pressure were closely related under a given compression load. The correlation appears as a linear relation with a slope "s" in log co-ordinates. This has allowed a dimensionless tightness parameter Tp to be defined.

    =

    Lrm

    *Lrms

    *P

    PTp

    where:

    P = absolute fluid pressure (MPa).

    P* = reference pressure (0.1013 MPa).

    Lrm = leak (mg/s) for a gasket with OD 150 mm.

    Lrm* = unit leak (1 mg/s) for a gasket with OD 150 mm

    s = slope of graph Lrm-P in log co-ordinates. Values raised to 0.5 for simplification purposes.

    Tp = 1 corresponds to a leak 1mg/s under atmospheric pressure for a gasket with OD 150 mm. The greater the Tp, the greater the gasket tightness.

  • Le December 15th 2005 AT n 790730/6J1/a Page 6/12

    If we plot the stress applied on the gasket vs. the Tp obtained during both ROTT tests on a log-log scale graph, we can represent the ideal behaviour of the gasket by means of straight lines (Figure 5: Idealized Sg-Tp graph)

    where:

    "Gb" = stress at Tp = 1 associated to the linear regression of the data in part A (loading).

    "a" = the slope obtained by the linear regression of the data in part A (loading). The combination of constants "Gb" and "a" characterises the seating stress and gives an indication of the capacity of the gasket to ensure the tightness.

    "Gs" = stress at Tp = 1 associated to the linear regression of the data in part B (unloading). It re-presents the capacity of the gasket to maintain the tightness after the pressure has been applied and in service, and its sensitivity to unloading.

    The definition of two other major parameters should also be noted:

    Tpmax(A) = Maximum value of the tightness obtained in part A (tightening) of ROTT test (out of the CRUSH part of the test)

    The greater the Tpmax, the greater the tightness on tightening.

    Tpmin(B) = Minimum tightness value obtained in part B (loosening) of ROTT test (out of the CRUSH part of the test).

    The greater the Tpmin, the greater the tightness on loosening

    Figure 4: Typical ROTT Sg-Tp graph

  • Le December 15th 2005 AT n 790730/6J1/a Page 7/12

    Figure 5: Idealized Sg-Tp graph

    2.2 CRUSH test procedure

    The CRUSH test is made as a continuation of the ROTT test. After having restored stress level S1, the CRUSH test consists in the cyclical application of gradually increasing compression loads on the gasket up to the maximum force for the rig. The leak is measured at each stress level under a 27.5 bar helium pressure as the test progresses. As the test is not aimed to quantify the leak but to check the gasket tightness throughout the stress cycles, the holding time for measuring the leak does not exceed 15 minutes. The test procedure is presented on Figure 6: CRUSH Test sequence

    The CRUSH test allows to determine the maximum allowable stress on the gasket before tightness is damaged. It consists in checking that the tightness is still present at stress level S1 (7 MPa) during successive loading and unloading cycles. The maximum allowable stress is reached when the leak seen at the stress level is greater than the leak measured at the initial stress level S1.

    Figure 6: CRUSH Test sequence

  • Le December 15th 2005 AT n 790730/6J1/a Page 8/12

    3. TESTED SPECIMENS

    The test has been performed on the following specimen

    Gasket type Gasket Dimension Testing temperature

    Novus Sealing Uniflon 50 gasket

    OD : 5.875

    ID : 4.875

    Thickness: 1/16

    Room

    Photo 1: Sample before test

  • Le December 15th 2005 AT n 790730/6J1/a Page 9/12

    4. TESTING EQUIPMENT AND MEASUREMENT INSTRUMENTS

    4.1 Test rig

    Tests are performed at CETIM using a compression machine or the ROTT test rig realized by TTRL.(Tightness Testing and Research Laboratory)

    Photo 2: Test rig

  • Le December 15th 2005 AT n 790730/6J1/a Page 10/12

    4.2 Gasket metrology The gasket is measured using a DIGIMATIC calliper (manufacturer: MITUTOYO), step value: 0.01 mm, capacity 300 mm 0.03 mm.

    4.3 Leak measurement Leaks < 0.018 mg/s are measured using an ASM 180T type helium mass spectrometer (CETIM reference: SPE005).

    The helium mass spectrometer is calibrated each week.

    4.4 Compression plates Gasket is compressed between two dimensionally stable plates:

    Testing plates

    Dimension 43x88 and 114x157 mm (sealing face, raised faces)

    Surface condition Ra = 250 AARH (6.3m)

    4.5 Gasket crushing measurement Crushing is measured from 2 sensors set 180 apart on the outside of the compression plates.

  • Le December 15th 2005 AT n 790730/6J1/a Page 11/12

    5. RESULTS

    5.1 ROTT Test results - gasket constants

    Computation of ROTT test on Novus Sealing "Uniflon 50" gasket leads to following gasket constants.

    5.2 CRUSH Test result